Joby Explains Its Unified Controls; Archer Publishes Less Cockpit Detail

Both aircraft need computers to blend hover, transition and cruise. The public difference is how much each company has explained about the pilot interface, not digital versus traditional control.

Neither aircraft is mechanically conventional

A pilot cannot directly coordinate six tilting propellers—or twelve engines, six tilting axes and aerodynamic surfaces—through cables and a helicopter-style set of independent controls. Both aircraft use fly-by-wire. Pilot commands go to computers, which interpret the desired motion and allocate it among motors, actuators and control surfaces while enforcing the approved envelope.

Describing Archer as “traditional” can therefore mislead. Its public certification page confirms redundant fly-by-wire flight controls and continuously monitored critical systems. What Archer has not published in equivalent detail is the exact production cockpit interface and control-law philosophy. Absence of public detail is not evidence of simpler or older mechanics.

Joby’s unified-control concept

Joby calls its system unified flight control. Public demonstrations describe one inceptor for aircraft motion and another control for acceleration and deceleration, allowing the flight computers to manage propulsion tilt and surface blending. The aim is for a fixed-wing commercial pilot to command where the aircraft should go without manually changing control technique at transition.

Joby lists a triple-redundant flight-control computer and dual critical actuation systems. Its 2023 pilot-on-board campaign evaluated vertical takeoffs, acceleration, partial transition, runway tracking and vertical landings. Later flights completed full pilot-on-board transitions. These tests support handling development; certification testing defines the final approved behavior and failure cases.

Archer’s controller has a different allocation problem

Midnight’s fly-by-wire system commands twelve engines in hover, tilts the forward six, reduces and stops the aft six in wing-borne cruise, and blends aerodynamic surfaces as airspeed rises. Six battery packs power diagonally paired forward and aft engines, so the control system also has to preserve controllability after a pack or engine fault.

That architecture does not imply a conventional cockpit. It implies a complex control allocator hidden behind whatever inceptors Archer certifies. Archer says critical flight controls are operationally redundant and that the aircraft can complete flight after an engine or battery-pack shutdown. The precise cockpit actions and degraded-mode limits are not provided on its public overview.

Computer count is only one layer

Three computers can vote, monitor or divide functions in several ways. The number does not reveal processor independence, software partitioning, sensor paths, power separation or actuator redundancy. Likewise, saying “redundant fly-by-wire” does not specify how many lanes exist. Certification examines common-cause failures and independence, not just duplicated boxes.

The system includes air-data, inertial and position sensors; communication networks; motor controllers; tilt actuators; aerodynamic-surface actuators; displays and pilot controls. A shared requirement or erroneous sensor can affect multiple nominally redundant channels. Design assurance, monitoring and graceful degradation determine whether redundancy is useful.

Pilot workload is an operational specification

Unified controls can lower training burden by making vertical and wing-borne phases feel like one task. They can also hide automation state unless displays and alerts make the aircraft’s mode and remaining capability clear. Pilots need to understand what the system will do after a motor, battery, actuator or sensor fault—especially during transition near the ground.

A fair comparison would use representative pilots in simulators and aircraft across normal, abnormal and emergency scenarios. Measure tracking error, response time, control reversals, mode confusion and workload. Include gusts, rejected transitions, missed approaches and degraded propulsion. “Intuitive” is a testable human-factors claim, not a substitute for results.

Certification and training close the loop

The FAA’s powered-lift framework addresses pilot qualification and operations, while type-specific training must reflect each aircraft. Joby is developing simulators with CAE and has a Part 141 flight-school certificate. Archer also holds pilot-training and operator certificates. Those organizational approvals do not certify the aircraft design, but they show both companies preparing the training system around it.

Automation must remain legible

A well-designed controller does more than make normal flight easy. It tells the pilot when capability has changed and makes the next safe action obvious. Mode annunciation, tactile behavior, alerts and checklist design are therefore part of the engineering comparison. If a propulsor or tilt actuator becomes unavailable, the pilot needs a stable command relationship rather than a lesson in control allocation.

That standard can be tested without access to source code. Put crews through repeated simulator scenarios, change failure timing, and measure whether they recognize the condition, maintain the flight path and choose the approved landing option. The design that produces fewer surprises under stress has the better human interface, regardless of how familiar its sticks look.

Training records can then show whether that advantage persists across pilots.

Bottom line

Joby has made its unified-control idea and triple-computer architecture unusually visible. Archer confirms redundant fly-by-wire while disclosing less about cockpit mechanics. Both depend on software-mediated control; certified handling, failure behavior and pilot workload will be the meaningful comparison.